To power fist or not?

[Ominous Anonymous]

It's the same reason a twin-linked Assault Cannon is better than a Lascannon for popping Land Raiders. Sure, the MotW is random and doesn't have the rerolls (not by itself, anyways) but feels more reliable.

[Valerian]

As a Blood Angels player I'd say Rending is underrated, especially against Dreads.

 

Yes, but as a Blood Angels player you can get a whole lot of Rending attacks together, which averages out to a good outcome. A whole Death Company charging a Dreadnought is quite a bit different than a single MotW Hunter in a Pack.

 

The Power Fist is nice but with only one attack and against most armies you'll have a 50% chance to miss it per turn.

 

Okay so Hunter with Fist gets 2 attacks on Charge/Countercharge with 50% chance of miss leading to an average 1 attack that will get to roll for either a Wound or Armour Penetration. Chances are great for a Wound with no Armour Save possible, that will Instant Death most things in the game (at about 83%) and okay for rolling on the Damage table for an attack against a Dreadnought (50% for either Glancing or Penetrating hit). Not great, but it could be worse.

 

Lets look at the same guy with MotW. He'll get an average of 5.5 attacks on the Charge/Countercharge, so lets round up to an even 6. Of those, he, too, has a 50% chance of a miss leading to an average 3 attacks that will get to roll to Wound or for Armour Penetration. Of those three, he will average only 1.5 Wounding Hits, which will usually be as effective as a Chainsword, with no possibility of Glancing or Penetrating a vehicle, and allowing Armour Saves for the enemy. For each of the 3 average rolls to Wound or Penetrate, there is a 17% chance per roll (51% overall) {this part is not accurate; I don't know why I wrote that, I know better - I screwed this part up, apologies.} that a Rend is achieved which would cause a Wound with no Armour Save allowed; this probability is significantly less than that of our Power Fist, and has no possibility of causing Instant Death on anything more significant than a 2 Wound Grot ;)

As far as Armour Penetration goes, the Rending attack has a 66% for rolling on the Damage table for an attack against a Dreadnought, which is good, but still less likely to be achieved than with the Power Fist (by about 16%).

 

Admittedly, the chances change considerably in the 2nd and following turns of combat, as the drop of 1 attack to the Power Fist model reduces his effectiveness in half, whereby it has very little effect to the attacks of the MotW guy.

 

V

 

Ninja'd by Grey Mage - took too long to type while trying to work at the same time... results are the same however, Power Fist is on average better. MotW may give the possibility of doing much better, with the mulitple Rending hits scenario, but over the long term it is just as likely to go several turns with absolutely no Rending hits.

 

Edit: Had to go back and review what I had done, based on feedback that I had screwed up. I certainly had, but only for one small portion of the overall assessment which I highlighted above. End result is still the same that the odds for effecting Armour 12 vehicles and Wounding MEQ are with the Power Fist over the MotW guy.

[Freman Bloodglaive]

Please remember that probabilities don't add, they multiply.

 

Three 17% chances of wounding are not a total of 51% they multiply to give 1-(1-p)^3 = 43%

 

For MOTW against armour 12.

 

Ignoring number of attacks for the moment.

 

1/2 to hit, 1/6 to roll 6, 2/3 of rolling 3+ to do some damage. Total 2/36 or 1/18.

 

Power fist.

 

1/2 to hit, 1/2 to do some damage. Total 1/4.

 

Q1 = 1-(1/18)=17/18

Q2 = 1-(1/4) =3/4

 

Now the number of attacks come in. We need to figure out how many attacks are necessary to drop 17/18 to 3/4

 

2 attacks (17/18)^2 = 0.89

3 attacks (17/18)^3 = 0.84

4 attacks (17/18)^4 = 0.79

5 attacks (17/18)^5 = 0.75

6 attacks (17/18)^6 = 0.71

7 attacks (17/18)^7 = 0.67

8 attacks (17/18)^8 = 0.63

 

The answer is 5, although I did the rest of them for completeness. Remember that these are probabilities of not damaging, subtract them from 1 to find the probability of damage occurring.

 

Since an average dice roll is 3.5 [(1+6)/2] and I'd truncate rather than round up to 3.

 

On the charge with a MotW model you'll roll on average 5 attacks. That is on a par with a single power fist attack. Of course on the charge the power fist has two attacks.

 

2 attacks (3/4)^2 = 0.56 and that is better than anything that the Wulfen can roll.

3 attacks (3/4)^3 = 0.42 for the Wolf Guard leader.

5 attacks (3/4)^5 = 0.24 on the charge a pack leader and hunter with power fist have a 76% probability of doing some damage to the dreadnought.

 

Powerfist is the way to go really. MotW may be fluffy but it's not combat effective.

[Grey Mage]

Its not even fluffy.

[Valerian]

Please remember that probabilities don't add, they multiply. Three 17% chances of wounding are not a total of 51%

 

You are absolutely correct, I made a common statistical error, and I even knew better than to do that. Good catch and thanks for correcting me. I know what I was trying to convey, but did it very poorly. I won't bother trying to re-explain the outcome that I meant to describe, as you already did that, and did so much more clearly than I could. As you demonstrated, the odds are with the Power Fist against both infantry MEQ and against enemy Dreadnoughts over MotW.

 

V

[megamat008]

For MOTW against armour 12.

Do you know that there isn't a single vehicle in the game with Rear AV12 ? The vast majority of them have Rear AV10, so if you roll a 6 to penetrate you'll auto penetrate.

 

So on a vehicle moving less than 6", you'll average 0.46 penetrating hits on the charge with MotW, while a PF will average 0.66 penetrating hits.

 

The difference is minimal, while MotW is cheaper and better vs infantry.

[Grey Mage]

For MOTW against armour 12.

Do you know that there isn't a single vehicle in the game with Rear AV12 ? The vast majority of them have Rear AV10, so if you roll a 6 to penetrate you'll auto penetrate.

 

So on a vehicle moving less than 6", you'll average 0.46 penetrating hits on the charge with MotW, while a PF will average 0.66 penetrating hits.

 

The difference is minimal, while MotW is cheaper and better vs infantry.

So you never face Dreadnaughts, Armored Sentinels, Defilers, Soulgrinders, or Ironclads???

 

Who are your opponents and why dont they have any tactical acumen?

[Valerian]

For MOTW against armour 12.

Do you know that there isn't a single vehicle in the game with Rear AV12 ? The vast majority of them have Rear AV10, so if you roll a 6 to penetrate you'll auto penetrate.

 

So on a vehicle moving less than 6", you'll average 0.46 penetrating hits on the charge with MotW, while a PF will average 0.66 penetrating hits.

 

The difference is minimal, while MotW is cheaper and better vs infantry.

 

MotW is not better vs. infantry, didn't you see that part?

 

V

[megamat008]

For MOTW against armour 12.

Do you know that there isn't a single vehicle in the game with Rear AV12 ? The vast majority of them have Rear AV10, so if you roll a 6 to penetrate you'll auto penetrate.

 

So on a vehicle moving less than 6", you'll average 0.46 penetrating hits on the charge with MotW, while a PF will average 0.66 penetrating hits.

 

The difference is minimal, while MotW is cheaper and better vs infantry.

So you never face Dreadnaughts, Armored Sentinels, Defilers, Soulgrinders, or Ironclads???

 

Who are your opponents and why dont they have any tactical acumen?

Well I try to never get my GH in combat with Walkers, despite having very good opponents, including some of the best ranked in Europe

 

http://www.tabletoptournaments.net/fr/t3_e...gid=3&cid=0 ;)

 

Most of the Walkers are either slow or easily shot after disembarking from their DP.

 

The only really threatening Walker is IMHO the Soulgrinder (I almost always run 3 of them in my Daemon army) which can not be dealt with even with two PF in a GH squad, so I'll leave this job to Lone Wolves or Wolf guards.

 

Generally speaking my GH try to stay as long as possible in their Rhino, or if I DP them I try to destroy the aformentioned targets with double melta.

 

MotW is not better vs. infantry, didn't you see that part?V

 

Against Orks you'll average 1,15 kills with MotW on the charge, and only 0,83 with a PF.

 

Against IG MotW will kill 1,63 guards, and a PF 1,11.

 

PF is only better against 3+ or better save opponents, but once again that is not your GHs duty.

[Valerian]

MotW is not better vs. infantry, didn't you see that part?V

 

Against Orks you'll average 1,15 kills with MotW on the charge, and only 0,83 with a PF.

 

Against IG MotW will kill 1,63 guards, and a PF 1,11.

 

PF is only better against 3+ or better save opponents, but once again that is not your GHs duty.

 

I should have not generalized. MotW is better against enemy infantry with no armour saves (Orks and IG) - but hell, against those types of troops a regular chainsword works just fine, so you don't need any type of special close combat weapon and your 15 points is better spent on an extra Marine. PF is better against 4+ and better personal armour saves, which is really all that is relevant, so once again the Fist wins out. I don't see why you would say that fighting enemy infantry is not your Hunters' duty. It seems killing other troops is a large part of what Hunters do, and are very well suited for.

 

V

[maznaz]

True, and maybe I wasn't clear about it, but the 51% wasn't the chance of Rending, but the "average total of Rends achieved". In other words, needing a roll of a 6 to Rend, and 3 Hits available to try, you will average out to 1 Rend for every 2 combat rounds, or .5 per turn. That's what the .51 was describing in my example above. I probably could have been more precise in my use of terminology/notation, but the end result is the same.

 

No that's incorrect. As he said, the probabilities don't add up, they multiply. The chance of rolling one six in six rolls is not one. The chance of rolling one six in three rolls is not 1/2.

[Marek Grimfang]

Rank and file never get expensive upgrades, except an old GH pack I have from 3rd Ed.

 

WG with PF, and 2x GH with PF, Melta Gun.

 

This was what I refered to as the "Can opener" squad.

[jimmyjames2112]

I think the fist is a nessesary tool if you dont have a WG with a fist in the squad. there are too many independents and other multi wound models that die to a S8 attack to not justify having it. even if you face IG or Nids alot. Fists are almost a sure fire way to pen a tank where you need a 6 on a MotW to get a pen. Try rending down Biker nobs even, Ill take the fist in every mobile squad every game. feel free to take MotW and a fist if you need the extra punch from your squad. but defiantly take a fist. That said.. a fist is a better option on a WG with the unit.. if you LRR/LRC you can get HQ, a WG and a full 10 man GH squad and get all the benefits. IMO power swords just arent very good VS t4 or better. 2 attacks. 1 hit, only a 50% kill ratio for 15 points. ANd thats assuming they dont just pile the power weapon wounds on the same model and really reduce the effectiveness

[Grey Mage]

Well I try to never get my GH in combat with Walkers, despite having very good opponents, including some of the best ranked in Europe

 

http://www.tabletoptournaments.net/fr/t3_e...gid=3&cid=0 :)

 

Most of the Walkers are either slow or easily shot after disembarking from their DP.

 

The only really threatening Walker is IMHO the Soulgrinder (I almost always run 3 of them in my Daemon army) which can not be dealt with even with two PF in a GH squad, so I'll leave this job to Lone Wolves or Wolf guards.

 

Generally speaking my GH try to stay as long as possible in their Rhino, or if I DP them I try to destroy the aformentioned targets with double melta.

Ahh, I see... we have very different schools of thought on what our GHs duty is.

[Ominous Anonymous]

Its not even fluffy.

 

Maybe not for Space Wolves...:)

 

Anyways, I'm taking it for the same reason other people take it: it's as cheap as a power weapon but allows me to deal with armor in a maxed-out Grey Hunter squad in a rhino.

[WG Vrox]

basically anyone who takes it for it's rending in place of a power weapon or power fist is going to be disappointed. Take it in addition to not instead of.

 

Only time I would consider taking a lone Motw model would be on a WS pack, running 5 WS MB MG and MotW in an effort to keep them cheap and effective @ 125

 

And even less so but still maybe on a foot bound rear guard GH pack to have a chance to deal with any surprises, but a WG with CML is the better option for that.

[Grogugluk]

I still don't think an 11% chance to kill a dread on the charge will justify that many points.

 

I guess it again completely depends on the squad as a whole. If you have no WG then its probably an ok idea to keep them more safe.

 

If you do have a WG in the squad I think its a waste of points.

 

I also think that having a wulfen and a banner is a much better way to spend those points and I think math wise it would show it also.

 

 

 

I also think one of the big things against the fist is that its only even decent in the first round of combat. After that its not very good at all.

 

Maybe everyone fights squads that they destroy in the first round but against big ork mobs, plague marines, other elite marine squads almost all the time you are going to have a second round of CC where that 1 Attack powerfist is terrible for the points.

[the jeske]

I still don't think an 11% chance to kill a dread on the charge will justify that many points.

how is instant killing meq IC without eternal warrior ? or negating buffs for being on a bike or a pm. instan killing nobs . its not just being able to hurt WL/dreads that makes the fist better.

even against meq you hit on+4 and then wound on +4 . with a fist you have one attack less but wound on +2.

 

 

all the time you are going to have a second round of CC where that 1 Attack powerfist is terrible for the points.

yes but that one attack , if it hits has a huge chance to kill. a pw will have one more attack , but wounding on +4 against nobs or +5 against pm will make it hard to actually kill a model.

[ChapniK]

thank you for the math

 

 

But then again, I hate mathhammer- Fists are better because they dont require you to be a crazy doomed beserker in the middle of your friends, because they always ignore armor saves, and because I have so many of them sitting around I could make a chaosspawn of solid powerfists and still equip all my squads.

@ grey mage Well i have a new signature

[Grogugluk]

I'm not saying powerfists aren't amazing. I think they with all the dreads and the ability to one shot a lot of HQ's you should have a bunch.

 

In my list I think I'm using around 5 of them on wolf guard with combi weapons. I think they are a really nice staple I'm just saying that for grey hunters basic trooper they aren't worth it.

 

 

Would you pay 50 points for one? I'm not taking either of the power weapon options on my Grey Hunters.. I don't think its worth it. I'd rather spend the points somewhere better.

 

I think the argument is whether its a good use of points to outfit your Grey Hunters with powerfists on the trooper.

[Valerian]

True, and maybe I wasn't clear about it, but the 51% wasn't the chance of Rending, but the "average total of Rends achieved". In other words, needing a roll of a 6 to Rend, and 3 Hits available to try, you will average out to 1 Rend for every 2 combat rounds, or .5 per turn. That's what the .51 was describing in my example above. I probably could have been more precise in my use of terminology/notation, but the end result is the same.

 

No that's incorrect. As he said, the probabilities don't add up, they multiply. The chance of rolling one six in six rolls is not one. The chance of rolling one six in three rolls is not 1/2.

 

Editing for clarity:

 

You are right, of course. I have apparently done a horrible job of trying to convey an idea here, so I apologize for that. I'll try one more time. The actual chance or probability never changes, and will remain at 16.7% no matter how many times you make the attempt from 1 to Infinite tries to roll a 6 on a single roll of a die; I think we are all in agreement here, correct? The chance of rolling one six in six rolls is indeed not one; I completely agree with that statement.

 

However, let's look at it another way, and see if you and I agree on this:

 

Let's say that we are going to roll our d6 a total of 600 times and record the result each time. With equal chances of each of the six sides/faces landing facing up, we would expect to have results where we rolled a 6 about 100 times out of our 600 attempts (likewise we would expect about 100 rolls of each of the other faces having occurred, too). Now, there is no guarantee for these results, as each roll is discrete, and the die has no "memory" of previous outcomes, but we still expect that about 1/6 of our attempts to land on one designated face (in this case the '6', because achieving the Rend is what is important). Are we in agreement so far?

 

Okay, now lets say that our 600 rolls above were done over the course of about 200 Combat Turns (many years of gaming with this army list, eh?) in which our MotW model was a participant. Based on the above, we would expect the MotW model to achieve about 100 Rends (a result of a 6 on the die) out of his 600 tries, and thus that he would get an average of a single Rend for every two combat phases (100/200) over the long term.

 

Please let me know if this is not an appropriate way to think about this problem.

 

V

[Skrubb]

Did you not at all read what you quoted? I just said that the .5 or 1/2 wasn't the "chance of rolling a six in three rolls" it was the average number of Rends that you would get when rolling 3 attempts at trying to achieve a 6. In other words, over the long term, you would be likely to achieve a single Rending hit for every two Turns of combat - which is absolutely a correct way to look at it.

 

I am sorry for reviving this thread only to disclose myself as a pedant and a nerd, but in the Holy Name of Statistics...

 

I am afraid you are wrong, Valerian. An AVERAGE of 0.5 hits is not at all the same as "over the long term, you would be likely to achieve a single Rending hit for every two Turns of combat". There is namely a chance of scoring MULTIPLE rending hits as well. If the average is 1 hit, the median number of hits is zero.

 

If there is a one in two million chance of winning one million pounds you could have stated "over the long term, you would be likely to win 1 pound for every two draws of the lottery". This can be true or untrue depending on what you mean. Can NOT expect to win every second draw.

 

This is not an important difference if all you care about is the average, but as it is, one rending hit in each of four turns is much better than four rending hits in one and the same turn.

[Valerian]

Did you not at all read what you quoted? I just said that the .5 or 1/2 wasn't the "chance of rolling a six in three rolls" it was the average number of Rends that you would get when rolling 3 attempts at trying to achieve a 6. In other words, over the long term, you would be likely to achieve a single Rending hit for every two Turns of combat - which is absolutely a correct way to look at it.

 

I am sorry for reviving this thread only to disclose myself as a pedant and a nerd, but in the Holy Name of Statistics...

 

I am afraid you are wrong, Valerian. An AVERAGE of 0.5 hits is not at all the same as "over the long term, you would be likely to achieve a single Rending hit for every two Turns of combat". There is namely a chance of scoring MULTIPLE rending hits as well. If the average is 1 hit, the median number of hits is zero.

 

Okay, just trying to make sure I understand your issue. Are you saying that I should not have used the terminology of "average" to describe what I was trying to? No doubt I very easily could be misusing the term, after all, it has been 16 years since my last Statistics course. Please take a look at the post just before yours, in which I just did some editing in hopes of doing a better job of relaying the information that I have been trying to all along.

 

This is not an important difference if all you care about is the average, but as it is, one rending hit in each of four turns is much better than four rending hits in one and the same turn.

 

This is interesting; why would you claim that a single Rending hit in each of four turns is better than four Rending hits in a single (the same) turn? I think a statement like that depends very much on what assumptions that you are making about the battle. There are lots of cases where I would much prefer the four Rending hits at once (if I were to get to 'choose' between the two).

 

Regards,

 

V

[Skrubb]

Okay, just trying to make sure I understand your issue. Are you saying that I should not have used the terminology of "average" to describe what I was trying to? No doubt I very easily could be misusing the term, after all, it has been 16 years since my last Statistics course. Please take a look at the post just before yours, in which I just did some editing in hopes of doing a better job of relaying the information that I have been trying to all along.

 

I love this! :)

Your use of the term average is fine. It is your use of "every two turns of combat" that is disputable. There is LESS than 50 percent chance of achieving the rend. Freman Bloodglaive's calculations shows that you will achieve one or more rends only every FOURTH round of combat. In AVERAGE you will have one rend per TWO rounds of combat. How vital this distinction is, rest on the below.

 

This is not an important difference if all you care about is the average, but as it is, one rending hit in each of four turns is much better than four rending hits in one and the same turn.

 

This is interesting; why would you claim that a single Rending hit in each of four turns is better than four Rending hits in a single (the same) turn? I think a statement like that depends very much on what assumptions that you are making about the battle. There are lots of cases where I would much prefer the four Rending hits at once (if I were to get to 'choose' between the two).

It will of course depend on the specific situation, but generelly, by achieving four rends in one turn, you risk killing the machine four times over, which is a wast. If you achieve one rend in each of four turns, there is no risk of waste. A parallell: It is better to force saving throw frow two different models than two from one model. (Unless the specific situation is special, for instance in battel with a multi-wound model.

 

I will have to add: I respect Grey Mage and others for their fluff-based armies. I strive for that myself. I love the fluff. But I am also an economist with love of statistics, so it is very hard for me NOT to tend to mathhammer. But even so, I am not a very competitive player. I am a crap player who play for fun and have no plans of entering tourneys. Acually, I don't even play much. I mostly paint. In fact I use mathhammer to determine which models to paint! :P

[Grey Mage]

I will have to add: I respect Grey Mage and others for their fluff-based armies. I strive for that myself. I love the fluff. But I am also an economist with love of statistics, so it is very hard for me NOT to tend to mathhammer. But even so, I am not a very competitive player. I am a crap player who play for fun and have no plans of entering tourneys. Acually, I don't even play much. I mostly paint. In fact I use mathhammer to determine which models to paint! :)

Wich is funny, because I am a competitive gamer- I enjoy the utilization of tactics on the field to win against my opponents, who are also my friends. I however disdain mathhammer because as has been stated- dice dont care what they rolled last time, and I only care what they are rolling now.

 

While Mathhammer can, and in some cases should, be used to show trends in how things will go it is far from gospel on the effectiveness of a unit. This is for several reasons:

 

1) Unaccounted variables. Such as what other units are on the field, the terrain setup of the board, special rules provided by other units, and frankly, luck. *note, Im a firm beleiver in luck. While dice might tend to roll accross their ranges Ive met people who for whatever reason roll their dice and get the least opertune numbers on a regular basis. While not a mythological force or anything nessecairily it is a factor that can not be accounted for*.

 

2) Most dice use a system of "pips" and as such are not semetrical. The weight on them is subtly, every so subtly, off balance. Most often it is often balance in favor of a six, as the 1 is slightly heavier. In any case its not a huge thing on a roll by roll basis, but when looking at the huge number of rolls involved with statistics it will show out in the end. This is never accounted for in mathhammer that Ive seen.

 

3) Percent deviation. I cant remember the last time I saw percent deviation in mathhamer style statistics, yes I find its one of the most important points for checking the validity and usefullness of such calculations.

 

4) Wide potential. Most of the time Mathhammer is done with a weapons abilities against a WS 4, T 4, 3+ armor save model. However that isnt that important in the long run- for instance it wont tell you how it does against a rhino, a Dreadnaught, a Wraithlord, or a Plaguemarine. And most people dont want to put anything near the time and effort required to put forth a real table- so why should I bother listening if they wont bother giving a full analysis?

 

5) Most importantly its not fun for me. Im very good with fractions and their use, and I dont shy away from decimals- I can find the "likely" outcome in my head midgame without pausing. But thats BORING. I know from years of experiance that Ive got about a 40% chance of killing that landraider with my meltagun if I shoot it, and no chance of killing it if I dont. I also know from experiance if I dont kill it and potentially hurt the Honor Gaurd inside then next turn they will pop out and eviscerate my squad. So I decide to take the chance and fire the gun- I dont care about the exact numbers, empericial evidence will do me fine.

 

That being said I have great respect for mathematicians- while Im firmly superstitious Im also rather scientifically minded and I respect people who speak the language of "pure" physics. I also have a great amount of respect for painters- as I myself am a wordsmith, and only over average talent when it comes to pigments.